Impact of Feedback in Flipped Learning on the Development of Soft Skills of University Students

Abstract

This study investigates the impact of flipped learning (FL) with and without teacher feedback on resilience and perceived professional competence among university students. This quasi-experimental study, conducted over three academic years, involved 255 s-year students (mean age = 20.47 ± 2.63; 60.4% female) enrolled in a Primary Education Teaching degree program at a Spanish public university. Participants were divided into two treatment groups: Feedback FL and Non-feedback FL. Validated scales—the Brief Resilience Scale (BRS) and the Perceived Competence Scale for Students (PCNS)—were used for pre- and post-intervention assessments. Both groups demonstrated significant improvements in resilience and perceived competence, with no statistically significant differences between the Feedback FL and Non-feedback FL groups. Data analysis included Mann–Whitney U tests for inter-group comparisons, Wilcoxon tests for intra-group comparisons, and 2 × 2 repeated measures ANOVA, which revealed no interaction effects (Time × Treatment) for either variable. These results suggest that FL alone fosters the development of transversal skills such as resilience and perceived competence, regardless of teacher feedback. The findings highlight the effectiveness of active learning models like FL in higher education for promoting essential competencies. Future research should address additional soft skills and employ larger, more diverse samples to further explore the role of feedback and innovative methodologies in enhancing FL’s impact.

1. Introduction

The onset of the COVID-19 pandemic accelerated the digital transformation of higher education, forcing teachers and students to quickly adapt to new teaching and learning modalities (Pecci et al., 2024). Among these, blended learning has significantly evolved in higher education due to its capacity to merge the benefits of face-to-face instruction with the advantages of digital or distance learning (Boelens et al., 2017). This instructional framework is defined by its emphasis on providing students with access to educational materials, such as notes, videos, and recorded lectures, before attending in-person sessions. This approach allows students to interact with the core content at their own pace. In this way, individualization in teaching is maximized by promoting deeper understanding during classroom activities (Thai et al., 2017). A specific variant of blended learning, flipped learning (FL), has emerged as a promising alternative, reconfiguring traditional classroom dynamics (Bergmann & Sams, 2012). This enables classroom time to be devoted to practical, collaborative, and problem-solving activities, fostering active learning and student engagement (Santiago & Bergmann, 2021a). This model is grounded in key pillars, including flexibility, intentional engagement, continuous access to content, and ongoing support, with feedback as a central element to enhance students’ self-regulation and conceptual understanding (Chou & Zou, 2020; Hattie & Timperley, 2007). Due to the numerous advantages it offers to the educational community, the flipped classroom has garnered growing interest in educational research (Bosch-Farré et al., 2024). Thus far, FL is one of the methodologies most closely linked to blended learning. Studies conducted by Thai et al. (2017), Hew et al. (2021), and Chen and Hsu (2022) assert that the synergy of these two approaches enhances student engagement and academic performance. Recent studies in physical education highlight that FL enhances learning outcomes, motivation, and practice time, particularly through the integration of technology and active learning strategies (Campos-Gutiérrez et al., 2021; Gil-Botella et al., 2021; Soriano-Pascual et al., 2022).

In 2018, Jon Bergmann and a group of experts updated the definition of FL, highlighting its ability to transform traditional teaching dynamics by allowing fundamental concepts to be worked on outside the classroom, freeing up class time for more interactive, student-centered activities (Santiago & Bergmann, 2021b). However, despite the growing body of scientific literature on the effectiveness of FL, questions remain about the specific mechanisms through which this model influences learning outcomes. In particular, the importance of teacher feedback has been emphasized as a key element for enhancing meaningful learning and skill development in students (Knoke et al., 2024), aiming to replicate an engaging and motivating learning experience linked to their interests and the skills they will later use in the labor market (Lai et al., 2018).

Additionally, the COVID-19 pandemic has led to a massive introduction of technology into teaching, making the shift towards the use of active methodologies urgent (Bosch-Farré et al., 2024), and consequently, student motivation has increased through the use of digital tools in Physical Education classes. Indeed, as conceived by its creators (Bergmann & Sams, 2012), FL goes beyond simply “watching videos” or completing pre-class tasks; it involves meeting a set of characteristics that distinguish it from any other analogous model (e.g., distance learning). One of the main advantages of FL is the ability to offer timely feedback to students in the collective space, following their prior work in the individual space at home (Santiago & Bergmann, 2021b). Hattie and Timperley (2007) pointed out that feedback is one of the most effective interventions for improving learning, as it guides students on what is expected of them, how they should proceed, and how to regulate their own learning. Subsequent studies have confirmed that feedback not only improves academic performance but also deepens students’ conceptual understanding (Finn et al., 2018).

Chou and Zou (2020) emphasized that the quality of feedback is not only defined by its content but also by its timeliness, asserting that it must be constructive, specific, and delivered at the opportune moment to maximize its impact. Similarly, Haughney et al. (2020) underscore the importance of integrating innovative and flexible tools that cater to the diverse needs and preferences of students, thereby optimizing the educational experience. Nicol and McCallum (2021) highlight that feedback not only corrects errors but also plays a fundamental role in developing self-regulation skills, which are essential for fostering autonomous and sustained learning. Collectively, these studies confirm feedback as an indispensable component for enhancing learning within the flipped classroom model. Nonetheless, although teacher feedback is an essential component of the effectiveness of FL, it remains an underexplored area. This lack of scientific literature suggests a need for more focused research, as noted by Bosch-Farré et al. (2024, p. 23), to examine the role of feedback in emerging models that provide active learning by the student. At the university level, recent reviews have explored the impact of FL, highlighting its relevance and growing interest (Bosch-Farré et al., 2024; Galindo-Domínguez & Bezanilla, 2019; Prieto et al., 2021). Student motivation was the most researched variable in this context, with generally positive results regarding FL’s effect.

FL has been shown to significantly improve student engagement and, consequently, learning and academic performance (Gosálbez-Carpena et al., 2022; Østerlie et al., 2023).

However, few studies have addressed its effectiveness in terms of acquiring transversal skills that facilitate their integration into the labor market (Lai et al., 2018) and even improve learning (Tseng et al., 2019). Recent studies in Physical Education have demonstrated that FL not only enhances transversal skills but also fosters motivation, practical engagement, and skill acquisition, contributing to improved learning outcomes and professional readiness (Campos-Gutiérrez et al., 2021; Ferriz-Valero et al., 2024; Fernández-Río et al., 2023). The need for a better-prepared population places higher education at the center of the agenda for skills training for employment, as it is the level at which these skills are developed at a more advanced professional level. The Organization for Economic Cooperation and Development (OECD) and the International Labor Organization (ILO) recommend that these qualifications be addressed at the educational level (Cornali, 2018). In Europe, following the Bologna Process and within the framework of the Europe 2020 Strategy, it was indicated that, in addition to increasing the number of graduates, they should be better qualified to respond to the ever-changing and complex needs of the labor market. This implies the inclusion of academic training in the development of interpersonal skills, transversal competencies, or soft skills, among other existing denominations. This acquisition of transversal skills, commonly referred to as “Soft Skills”, is increasingly valued in the labor and social sphere. These skills, which include effective communication, collaboration and teamwork, creativity, adaptability, resilience, etc., complement technical knowledge and are fundamental for success in an ever-evolving work environment (Cornali, 2018). In this regard, Soriano-Pascual et al. (2022) emphasized the effectiveness of tools like Edpuzzle^® in enhancing sports-related learning activities in secondary education. In this context, higher education faces the challenge of designing learning experiences that promote the holistic development of students, fostering both the mastery of disciplinary content and the acquisition of transversal competencies. Currently, resilience is considered a personal strength of great importance within the positive psychology model (Haktanir et al., 2016). It is understood as the ability to adequately cope with various conflicts and life situations that would otherwise lead to psychosocial and emotional maladjustment (Masten & Obradovic, 2006). Given its importance and transversal nature in both the professional and personal domains, it is necessary to have tools that help understand this construction and, above all, to develop its capacity.

Some authors observe that students following FL seem to excel in self-learning skills (Li et al., 2020; Talan & Batdi, 2020), problem-solving (Ge et al., 2020; Xu et al., 2019), teamwork (Chen & Hsu, 2022), communication competencies (Senali et al., 2022; Turan & Akdag-Cimen, 2020), time self-management (Hew et al., 2021; Oudbier et al., 2022), and creativity (Oudbier et al., 2022; Senali et al., 2022). Regarding perceived competence, studies by García (2018) indicate that, to successfully perform 21st-century tasks, it is essential to possess cognitive, as well as intra- and interpersonal skills. This includes the ability to achieve personal goals and to have competencies that enable us to influence the world around us. Similarly, García highlights that our socio-emotional skills, such as perseverance and decision-making, are linked to awareness, perceived competence, and the regulation of our emotions. Therefore, the objective of this study focused on the role of teacher feedback in the context of FL, exploring its impact on the development of competencies and personal skills in university students, particularly in resilience and perceived labor competence.

2. Materials and Methods

2.1. Research Design

This study was conducted over three consecutive academic years (2021–2024) in the context of an official course offered during the second semester (January to May) at a Spanish public university. The course, worth 6 ECTS credits, is part of the undergraduate degree in Primary Education within the teacher training program. The course is a core subject in the curriculum, and its structure aligns with the university’s academic calendar and teaching framework. A quasi-experimental design with convenience sampling was employed, involving two treatment groups (experimental and control), with pre- and post-intervention assessments. The experimental group implemented FL, in which students viewed 16 subject-related videos before class (individual learning phase). Following this preparation, they utilized the collective classroom space to receive high-quality teacher feedback, including detailed, reinforcing, and corrective feedback. In contrast, the control group also viewed the same videos prior to class but did not receive any teacher feedback during the face-to-face sessions. This absence of feedback in the collective space introduces a variation in FL. According to its definition (as detailed in the preceding section), feedback is an essential component for an intervention to be classified as FL. This key distinction between the groups allowed for the targeted investigation of the role of feedback within FL, particularly in relation to the development of students’ competencies and personal skills.

To test the formulated hypotheses, four classes were assigned to the “Non-feedback FL” group and another four to the “Feedback FL” group, using convenience sampling, but maintaining the same teacher, who had experience in implementing FL, for both groups. This control helped mitigate potential effects attributable to the teacher, ensuring consistent content and teaching style across treatments. Designs previously used in similar studies were considered (Thai et al., 2017, 2020). To conclude, it is important to highlight that this study was approved by the ethics committee of University of Alicante (UA-2020-09-02).

2.2. Participants

A total of 255 s-year undergraduate students in the Primary Education teaching degree program participated in the study (Mean age = 20.47 years; SD = 2.63) at University of Alicante (Spain), of whom 154 were female (60.4%) and 74 (39.6%) were male. Inclusion criteria for this research were: (1) being enrolled for the first time in the core second-year course of the Primary Education Teaching degree, and (2) belonging to one of the groups whose teacher was an expert in FL. A total of 145 students were excluded from the study for failing to meet any of the following exclusion criteria: (a) not regularly attending classes, i.e., less than 80% of all sessions (n = 21); (b) not adequately completing the questionnaires (n = 76); or (c) not signing the informed consent form (n = 48). The selection of participants was deliberate, based on the availability of students enrolled in the course who met the inclusion criteria. After their inclusion was confirmed, participants were randomly assigned to either the experimental or control group. This randomization ensured equitable distribution and minimized potential biases in the study’s outcomes.

2.3. Intervention Program

Hastie and Casey (2014) suggest that a rigorous intervention should include: (a) a detailed description of the curricular elements involved; (b) thorough validation of the intervention model; and (c) a precise description of the program’s context. These sections are described below as accurately as possible.

The intervention was implemented within the framework of the compulsory second-year course “Didactics of Physical Education”, part of the Primary Education Teaching degree. This course is worth six ECTS credits under the Spanish university system. Its main objective is to provide future Primary Education teachers with foundational knowledge on motor learning and sports. The course is structured around eleven topics, of which only three were selected for the intervention (see

Table 1. Summary of research design (curricular elements).

Content	Video Resource (min)	Session (~115 min)
Topic 3
Teaching Styles I	Conceptual Approach	6:01
	Traditional Styles	6:52
	Participatory Styles (Reciprocal Teaching)	4:08
	Participatory Styles (Small Groups)	1:47
	Participatory Styles (Microteaching)	2:52
Topic 4
Teaching Styles II	Individualizing Styles	7:18
	Socializing Styles	3:33
	Cognitive Styles: Guided Discovery	5:09
	Cognitive Styles: Problem Solving	2:21
	Creative Styles	3:33
	Conclusions	4:04
Total		26:38
Topic 6
Basic Physical Capacities	Physical Capacities in Primary Education	4:47
	Flexibility	6:40
	Strength	8:06
	Endurance	8:24
	Speed	4:24
Total		32:21
		8 sessions (~16 h)

). The course assessment was based on field practice (20%), demonstration of mastery of acquired knowledge (25%), competence in designing a learning situation in Physical Education (25%), and a written exam (30%).

The intervention program was implemented by three instructors with extensive experience in FL, accumulating between five and seven years of practice with this method and an average of nearly 10 years of experience in university teaching (9.5 ± 2.52 years). Each intervention was led by a single instructor responsible for their group, minimizing potential biases arising from the involvement of multiple professionals. To enable students to watch the assigned videos, the EdPuzzle platform (https://edpuzzle.com) was used. Unlike other options (e.g., YouTube or Vimeo), EdPuzzle allows the instructor to monitor students’ progress. Students also had free access to personal accounts through an intuitive, user-friendly interface, ensuring seamless access to the videos without technical difficulties. A fundamental aspect of FL implementation in this study is that EdPuzzle enables students to answer questions related to the videos and allows instructors to monitor who watched the videos and under what conditions. For instance, the platform provides data on each student’s viewing time, repeated sections, and whether they answered the questions correctly, supporting the development of this research. As detailed in Table 1, the intervention took place over eight sessions of 115 min each, distributed across four weeks, with a total duration of approximately 16 h. Both the “Non-feedback FL” and “Feedback FL” groups followed the same course content, used the same platform (EdPuzzle), and watched the same 16 videos, totaling approximately 80 min. These videos were recorded by a single instructor (the principal investigator) but were designed and supervised by all involved instructors (Figure 1). Additionally, both groups responded to the same 45 questions (open-ended, true/false, and multiple choice; Appendix A). In short, the difference between the groups lay solely in formative assessment, provided exclusively to the “Feedback FL” group.

The intervention was structured into four phases, with the only difference between the groups being the presence or absence of teacher feedback.

• Phase 1: Only applied to the group with feedback. Corrective and reinforcement feedback is provided. By analyzing the students’ responses during their viewing of the video prior to the session, the teacher focuses on the students’ incorrect answers and the sections of the video that were replayed most often (indicating that the group perceives difficulty in those topics).
• Phase 2: Peer teaching. In the group setting, the session begins with a peer-teaching dynamic involving groups of 5–6 students, guided by the teacher. Over 10 min, each group reviews the key aspects of the pre-session video. Subsequently, groups are randomly selected to share their conclusions with the rest of the class. Finally, the group with feedback receives teacher feedback. The teacher provides high-quality feedback related to the discussed concepts and group interventions, emphasizing common errors or areas where the group struggles most.
• Phase 3: Practical scenario or case study. Students (working in small groups) must solve a practical case presented by the teacher. During the activity, in addition to offering reinforcement and real-time corrections, the teacher provides high-quality feedback (only to the feedback group). For the non-feedback FL group, the activity is conducted autonomously. The teacher offers guidance, promotes reflection, and encourages self-assessment but does not provide feedback.
• Phase 4: Structured discussion. The teacher moderates the session, allowing each group to present their solutions to the practical scenario or case study. The objective of this discussion is to reflect on the shared responses and integrate knowledge, thereby achieving a deeper level of understanding. The teacher may offer specific recommendations and clarifications.

This detailed intervention illustrates how flipped learning (FL) differs from traditional methods. In a conventional classroom, students are typically introduced to theoretical concepts for the first time during a lecture, with minimal interaction or feedback from the teacher. In contrast, FL transforms this process by utilizing class time for active participation, collaborative problem-solving, and real-time feedback. This dynamic approach allows for misconceptions to be addressed immediately and fosters the development of higher-order cognitive skills. Additionally, as highlighted by González-Zamar and Abad-Segura (2022), FL prioritizes active participation and promotes competencies such as creative problem-solving and critical learning. Moreover, its effectiveness in transforming group spaces into dynamic, interactive environments is emphasized.

Below, a summary graphic of the design used in this research is presented (Figure 2).

2.4. Instruments and Variables

To measure resilience and perceived competence, two quantitative scales with a 1–5 response option, qualitative in nature, ordinal scale, and Likert type, were used. First, the Brief Resilience Scale (BRS), a specific resilience scale (Smith et al., 2008), includes 6 items with a 5-point Likert response format (1 = strongly disagree, 5 = strongly agree). Items 1, 4, and 6 are reverse-scored to control response bias. The Spanish version of the scale (

Table 2. Brief Resilience Scale (BRS).

Factor	Items
Resilience	After the difficult situations I have experienced in my studies, I have recovered quickly (*)
	When something bad happens to me in practical activities, it has been hard for me to return to classes.
	When I have stressful situations within my practical activities, I have had difficulty moving forward.
	Normally, I have barely had any problems in my studies overcoming difficult situations (*)
	I need quite a bit of time to get over unexpected events in my academic life.
	I don’t need much time to recover from a stressful situation resulting from my studies (*)

Note: (*) reverse scored.

) was used in this study (Rodríguez-Rey et al., 2016), with Cronbach’s alpha values above 0.80 in both pre- and post-tests, indicating high internal consistency.

Secondly, students’ perceived competence (PCNS) was measured using a scale designed to assess students’ perception of competence (Hu & Bentler, 1999). The scale includes 10 items rated on a 5-point Likert scale (1 = Not at all competent and 5 = Fully competent) and preceded by the phrase: “As of today, in your opinion, what is your level of competence in…”. The scale is adaptable for students from various scientific disciplines, aligning with the specific competencies set in their study programs (

Table 3. Perceived competence scale for students (PCNS).

Factor	Items
Perceived Competence	61. (…) use assessment in its pedagogical role, not just as a means of accreditation, as a regulatory element that promotes the improvement of teaching, learning, and self-development, acknowledging the need for continuous professional growth through reflection, self-evaluation, and research on one’s own practice.
	62. (…) understand that the educational process in general, and teaching and learning processes in particular, are complex. Recognize that teaching practices must improve, be updated, and adapt to scientific, pedagogical, social, and cultural changes. Value the importance of participating in innovation and research projects related to teaching and learning, and of introducing innovative proposals in the classroom.
	63. (…) understand the characteristics and conditions in which school learning occurs and identify how it can impact student development, fulfilling the role of tutor by guiding students and their parents. This includes fostering understanding and cooperation with families, taking into account diverse family contexts and lifestyles.
	64. (…) motivate and encourage students’ academic progress within the framework of holistic education and promote autonomous learning, based on the goals and content of each educational level, with positive expectations for student progress. This entails avoiding established stereotypes external to learning and developing strategies that prevent exclusion and discrimination.
	65. (…) foster a sense of responsible, critical citizenship in students, encouraging the collaborative creation of democratic coexistence norms and the collaborative resolution of problematic situations and conflicts. Be capable of analyzing social inequalities within the complex relationship between education and school, and the teacher’s role in either perpetuating or transforming them.
	66. (…) design and develop educational projects, programming units, environments, activities, and materials, including digital resources, that allow for curriculum adaptation to the diversity of students and promote quality in the contexts where the educational process occurs, ensuring student well-being.
	67. (…) assume the ethical dimension of teaching, acting responsibly, making decisions, and critically analyzing educational concepts and proposals stemming from research, innovation, and educational administration.
	68. (…) have the proficiency to effectively use expressive body resources in various practical environments.
	69. (…) possess the skills to design recreational physical-sport activity programs in social, cultural, and diversity-oriented contexts.

Note: Preceding phrase: “As of today, in your opinion, what is your level of competence in…”.

). Cronbach’s alpha values ranged between 0.69 and 0.81 across all factors, indicating good internal consistency.

2.5. Statistical Analysis

The statistical software SPSS version 28.0.0.0 (190) was used to perform all analyses. Descriptive statistics were calculated for each factor (mean and standard deviation). The Kolmogorov–Smirnov normality test was conducted, yielding non-normal distributions in all cases (p < 0.05). Next, an intra-group comparison (Wilcoxon test) was performed to analyze the differences between the pre-test and post-test. Finally, to test the hypothesis, a 2 × 2 repeated measures analysis of variance (ANOVA) was used to strengthen the analysis. The dependent variables were resilience and perceived professional competence. Time (before and after the intervention) was the intra-group factor, while group (Non-feedback FL vs. Feedback FL) was the between-subjects factor. Levene’s test was used to check homoscedasticity, Mauchly’s test for sphericity, and Box’s test for covariance matrix equivalence. All assumptions were met for the dataset except for data normality and Mauchly’s test (thus, multivariate contrasts were considered). Effect size was calculated using Microsoft Excel (Dominguez-Lara, 2018). This magnitude was considered small when values ranged between 0.1 and 0.3, medium between 0.3 and 0.5, and large if greater than 0.5 (Cohen, 2013; Coolican, 2017). A 95% confidence interval was calculated for differences, and the significance level was set at p < 0.05.

3. Results

3.1. Initial Differences Between Treatment Groups (Pre-Test)

Baseline characteristics of both groups are presented in

Table 4. Mean ± standard deviation of initial differences between groups using the Mann–Whitney U test (pre-test).

Variable	Entire Sample		Non-Feedback FL (n = 130)		Feedback FL (n = 125)		Z	Sig.	ES
	M	DT	M	DT	M	DT
Soft Skills (range 1–5)
Resilience	3.06	0.44	3.07	0.48	3.04	0.39	−0.251	0.802	-
Perceived competence	3.27	0.70	3.24	0.67	3.30	0.73	−0.251	0.802	-

Note: M = Mean; DT= Standard Deviation; ES = Effect Size.

, including statistics of these differences obtained through the non-parametric Mann–Whitney U test. In the pre-test, the groups showed similar initial values for resilience (Z = −0.251, p = 0.802, 95% CI [−0.11, 0.11]) and perceived competence (Z = −0.251, p = 0.802, 95% CI [−0.12, 0.12]).

3.2. Longitudinal Differences Within Each Treatment Group (Pre vs. Post-Test)

Table 5. Mean ± standard deviation of the intra-group comparative analysis using the Wilcoxon test.

Variable	Pre-Test		Post-Test		Z	Sig.	ES
	M	DT	M	DT
“Non-feedback FL” (n = 130)
Soft skills (range 1–5)
Resilience	3.07	0.48	3.48	0.61	5.714	<0.001	0.50
Perceived Competence	3.24	0.67	3.59	0.72	4.977	<0.001	0.44
“Feedback FL” (n = 125)
Soft skills (range 1–5)
Resilience	3.04	0.39	3.43	0.61	5.381	<0.001	0.48
Perceived Competence	3.30	0.73	3.72	0.76	6.155	<0.001	0.55

Note: M = Mean; DT = Standard Deviation; ES = Effect Size.

presents the results obtained from the application of the non-parametric Wilcoxon test. The results indicate that the “Non-feedback FL” group showed significant improvements in resilience (Z = 5.714, p < 0.001, ES = 0.50, 95% CI [0.30, 0.70]) and perceived competence (Z = 4.977, p < 0.001, ES = 0.44, 95% CI [0.24, 0.64]). Similarly, the “Feedback FL” group also improved significantly in resilience (Z = 5.381, p < 0.001, ES = 0.48, 95% CI [0.28, 0.68]) and perceived competence (Z = 6.155, p < 0.001, ES = 0.55, 95% CI [0.35, 0.75]).

3.3. Final Differences Between Treatment Groups (Post-Test)

Table 6. Mean ± standard deviation of final differences between groups using the Mann–Whitney U test (post-test).

Variable	Entire Sample		Non-Feedback FL (n = 130)		Feedback FL (n = 125)		Z	Sig.	ES
	M	DT	M	DT	M	DT
Soft skills (range 1–5)
Resilience	3.46	0.61	3.48	0.61	3.43	0.61	−0.923	0.356	-
Perceived Competence	3.66	0.74	3.59	0.72	3.72	0.76	1.716	0.086	-

Note: M = Mean; DT = Standard Deviation; ES = Effect Size.

shows the results obtained after the application of the Mann–Whitney U test. After the intervention, no significant differences were observed between the groups for resilience (Z = −0.923, p = 0.356, 95% CI [−0.11, 0.05]) or perceived competence (Z = 1.716, p = 0.086, 95% CI [−0.01, 0.14]).

3.4. Hypothesis Testing

No interaction effect (Time × Treatment) was observed in any dependent variable (Figure 3). In other words, no statistically significant longitudinal differences (pre vs. post) were found (p < 0.05) in the impact of the treatment (feedback vs. no feedback) during the development of FL in either group.

4. Discussion

The development of soft skills in higher education is increasingly recognized as a crucial aspect of preparing students for the demands of the modern workforce. These competencies complement technical knowledge and enhance employability and professional effectiveness (Santos et al., 2023). While numerous pedagogical approaches, including flipped learning (FL), have been implemented to foster these skills (Prieto et al., 2021; Bosch-Farré et al., 2024), the role of feedback in shaping soft skills remains underexplored. This study aimed to address this gap by analyzing the differential impact of feedback in the context of FL.

This study analyzed the effect of feedback in the FL approach on the development of soft skills, specifically perceived competence and resilience, for students in the Primary Education Teaching degree, where this study was applied. Despite the theoretical rationale supporting the inclusion of feedback as a key component in fostering these skills (Ryan & Deci, 2019), the results did not indicate significant differences between the Feedback FL Group and the Non-Feedback FL Group.

While both groups exhibited improvements in perceived competence and resilience over the course of the intervention, these gains were comparable regardless of the presence or absence of feedback. This finding suggests that the overall structure and active learning principles of the FL may have contributed to the observed improvements in soft skills, irrespective of feedback. Similar findings were observed by Ferriz-Valero et al. (2022a, 2022b), where FL promoted significant learning gains and motivation. In this context, the role of peer interaction, self-paced learning, and the active engagement encouraged by FL might be particularly relevant. These findings resonate with prior research suggesting that active learning environments, such as FL, inherently foster soft skills through mechanisms like peer collaboration and autonomy (Gilboy et al., 2015; Craft & Linask, 2020). While prior studies (Gilboy et al., 2015; Bosch-Farré et al., 2024) have demonstrated the effectiveness of FL for improving transversal competencies, this study specifically highlights that these gains occur even in the absence of structured teacher feedback, addressing a previously unexplored gap in the literature. The absence of significant differences may be due to the relatively short intervention duration and the quality or frequency of feedback provided. Research highlights that timely, specific, and actionable feedback is crucial for influencing complex competencies like resilience (Ryan & Deci, 2019; Zhang et al., 2023). Future studies should explore whether adjustments in feedback strategies—such as increasing its frequency, incorporating peer review, or leveraging technology—could amplify its impact within the FL approach.

In line with these observed improvements, the enhanced resilience and perceived competence correspond to several transversal competencies outlined in the White Paper for Teaching Degrees by the National Agency for Quality Assessment and Accreditation (ANECA, 2005). For example, adaptability to new situations (a systemic competence) and problem-solving (an instrumental competence) are reflected in the resilience evaluated here, while perceived competence aligns with competencies such as autonomous learning, organizational and planning skills, analysis and synthesis abilities, and effective oral communication—primarily instrumental competencies. It is noteworthy that, according to ANECA (2005), digital competence is among the five least valued by students, which highlights the importance of fostering a broad range of soft skills. Moreover, these findings resonate with UNESCO’s emphasis on key 21st-century learning skills—such as collaboration, communication, productivity, content creation, and personal attributes like initiative, resilience, responsibility, risk-taking, and creativity—further underscoring the relevance of developing such competencies in contemporary educational settings (García, 2018).

The absence of differences between groups does not diminish the importance of feedback in other areas of educational practice. Research has consistently shown its positive impact on variables such as academic performance and motivation (Zhang et al., 2023; Finn et al., 2018). However, these findings highlight the need for further investigation into the specific conditions under which feedback is most impactful in promoting soft skills. It is possible that the type, frequency, or perceived quality of the feedback provided in this study was not sufficient to yield measurable effects in these areas. Additionally, the relatively short duration of the intervention might have limited the potential for detecting long-term impacts on more complex competencies like resilience. Another consideration is that the instruments used to assess perceived competence, and resilience may not have been sensitive enough to capture subtle differences between the groups. Soft skills are multifaceted constructs influenced by a range of contextual and individual factors, and their development often requires sustained interventions and robust measurement tools (Santos et al., 2023).

Further supporting these results, studies such as that of Knoke et al. (2024) indicate that both intrinsic and extrinsic motivation can improve through non-traditional, digitally enhanced methodologies like gamification in physical education, bolstering a range of competencies similar to those evaluated here. Even in the absence of feedback, activating metacognitive learning strategies can enhance performance and facilitate autonomous learning—an instrumental competence highlighted by ANECA (2005). Likewise, Bosch-Farré et al. (2024) found that FL, compared to traditional ones, reduced dropout rates and strengthened competencies such as autonomous learning, problem-solving, teamwork, communication, time management, and creativity. These competencies, integral to the White Paper’s framework, align with the emphasis that the OCDE (2015) and SEL programs in countries like Mexico and the United States place on socio-emotional skills. Recognizing that these abilities are not fixed personality traits but can be developed over time (Duckworth & Yeager, 2015; West, 2016) underscores the potential of instructional models like FL for cultivating such skills. In line with this perspective, B-HERT (2002) recommends that teaching staff broaden their pedagogical strategies—employing simulations, case studies, and project-based activities—to serve as mentors, facilitators, evaluators, and role models for developing these generic competencies.

These findings hold valuable implications for teaching and learning, particularly in courses like physical education. Educators can confidently implement FL to promote critical competencies such as resilience and perceived competence, even when feedback is limited. However, to maximize the model’s impact, teachers could integrate regular, high-quality feedback—whether teacher-driven, peer-based, or technology-mediated—into FL activities. Combining FL with feedback practices may offer a balanced approach, fostering both skill acquisition and student engagement in diverse educational contexts. Educators could incorporate formative feedback strategies—such as peer reviews during collaborative activities or automated tools like online quizzes that provide immediate feedback—to enhance the learning process while ensuring that feedback is actionable and consistent.

Despite the valuable insights gained, this study presents certain limitations that warrant consideration. Another limitation of this study is the use of a single-institution sample, which may limit the generalizability of the findings to other educational contexts or institutions. Differences in institutional culture, teaching practices, and student demographics could influence the outcomes of similar interventions. Future studies should replicate this research across multiple institutions and diverse educational settings to confirm the robustness and applicability of the results. The use of a convenience sampling method may have introduced biases related to participants’ prior training or subject-specific knowledge. Future research should evaluate students’ digital proficiency and quantify the time and frequency of digital tool use. Additionally, broadening the range of assessed soft skills, including digital competence, informal learning abilities, second-language fluency, content creation, synthesis capacity, and time management—could provide a more comprehensive understanding of FL’s influence. Given that one of FL’s key objectives is to reduce theoretical lecture time and increase practical sessions, exploring the interplay between these skills and class dynamics would be highly informative. In addition, examining predictors of outcomes, such as self-esteem, which has been linked to resilience (Peña et al., 2020), or comparing improvements in learning with and without feedback using alternative pedagogical approaches (e.g., peer instruction, just-in-time teaching, team-based learning, or even integrating AI tools), as suggested by Prieto et al. (2021), may help clarify under what conditions feedback can most effectively enhance soft skills.

Therefore, future studies should consider employing a combination of qualitative and quantitative methodologies to gain a more comprehensive understanding of how feedback influences soft skills in diverse educational settings. Such approaches would allow for a deeper exploration of the nuanced effects of feedback—such as its perceived usefulness, timing, and delivery formation, and the development of soft skills within active learning frameworks like FL. As educational contexts continue to evolve, identifying the most effective strategies for integrating feedback with active learning and technological tools remain essential for supporting the development of the transversal competencies demanded by the contemporary world.

5. Conclusions

This study’s findings, consistent with its primary objective, indicate that the provision of teacher feedback within FL was not a decisive factor in fostering personal competencies such as resilience and perceived professional competence among university students. Both the feedback and non-feedback groups demonstrated significant gains in these competencies following the FL intervention, with no statistically significant differences observed between the groups. These results suggest that FL, irrespective of the presence of feedback, effectively contributes to the enhancement of transversal competencies in students.

These findings underscore the value of FL as a framework that inherently promotes active engagement and self-directed learning. Both groups showed meaningful improvements in perceived competence and resilience, highlighting the potential of FL to support the development of these critical competencies in higher education. Educators should therefore strive to balance the integration of feedback with other elements of the FL approach, ensuring a holistic approach to student development.

This study substantiates the potential of active learning models, specifically FL, to facilitate the acquisition of essential competencies in higher education, competencies which hold considerable relevance in both academic and professional environments. However, while feedback remains an instrumental pedagogical tool, its direct influence on the advancement of resilience and perceived competence within the FL framework appears limited.

This study contributes to the growing body of research on FL by demonstrating its effectiveness for fostering transversal competencies, while underscoring the need to refine feedback practices to optimize its impact on complex skills such as resilience and perceived competence.

Ultimately, by addressing an underexplored aspect of FL—the role of feedback—this study provides critical insights that can inform the design of more effective pedagogical strategies, fostering soft skills essential for success in higher education and beyond.